We are currently gathering data to determine the histologic and molecular differences in breast cancer brain metastases as a function of multiple fractionation regimens for WBRT using a xenograft mouse model. My collaborators have developed experimental models of brain metastasis of breast cancer using sequential rounds of intracardiac injection of tumor cells, harvesting of brain metastases, and culture. These models include the triple-negative cell lines MDA-MB-231 and 4T1, the Her-2+ cell lines Jimt-1 and BT474 and the Her-2 transfected cell lines MCF-7 and MDA-MB-231. Using the one month MDA-MB-231 model (231-BR), mice were treated with either a SRS or a WBR schedule e during the last two weeks of brain metastatic colonization. On day 14 post injection, WBRT was given as 15Gy in a single dose to one group of mice, compared to 30 Gy delivered in 10 equal doses over 10 days and ending on day 24 post-injection. These doses are considered to be radiobiologically equivalent for tumor cell killing potential (within a standard deviation of 1 Gy) calculated using an alpha-beta ratio of 10. Mice that received WBRT were compared to untreated controls. Initial results have demonstrated surprising differences. First, while both regimens significantly decreased brain metastasis formation, neither was completely effective, in keeping with clinical experience. WBRT reduced the population of micrometastases by more than 90%. In comparison, large metastases, comparable to a MRI-detectable lesion in a human brain, remained with an approximate 70% reduction. In addition, a prominent neuro-inflammatory response was identified in 231-BR metastasis using immunofluoresecence techniques with CD11B/CD45, and GFAP antibodies. The inflammatory effect persisted in those lesions which remained after radiation therapy. However, almost no inflammation was identified in brains without metastases that had received radiation treatment. Based on these exciting observations, we first aim to determine the effect of SRS and WBR schedules in multiple experimental brain metastasis models of breast cancer. The efficacy of these regimens will be quantified in vivo in three model systems. For the 231-BR model, I have also cultured tumor cells remaining after WBR from 5-8 mice per treatment group, and will retest these cultures by re-injection and WBR treatment to determine if an altered radiation sensitivity is detectable. The neuro-inflammatory response, identified by activated microglia and astrocytes surrounding and within metastases, and blood-brain barrier structure and patency, will be quantified. These experiments are expected to provide a phsiologically relevant understanding of radiation therapy efficacy in a breast cancer preclinical model. Secondly, we hope to identify gene expression alterations associated with metastatic persistence after radiation therapy. Tumor cells surviving after WBR in two models will be laser capture dissected, and the nucleic acids isolated, amplified and tested in microarray experiments. These samples will be compared to unirradiated controls. Differentially expressed genes will be confirmed at the mRNA and protein levels. These experiments should identify genes and pathways that mediate radiation resistance, for further experimentation.